Method of reconstructing three dimensional image using structured light pattern system

Abstract

A method of reconstructing a three dimensional image using a structured light pattern system is provided as follows. A class identifier of an observed pixel on a captured image by a camera is extracted. The observed pixel has a coordinate (x, y) on the captured image. A first relative position of the x coordinate of the observed pixel in a tile domain of the captured image is calculated. A second relative position of one of a plurality of dots in a tile domain of a reference image using the extracted class identifier is calculated. A disparity of the observed pixel using the first relative position and the second relative position is calculated.

What is claimed is: 
     
       1. A method of reconstructing a three dimensional image using a structured light pattern system, the method comprising:
 extracting a class identifier of an observed pixel on a captured image by a camera, wherein the observed pixel has a coordinate (x, y) on the captured image; 
 calculating a first relative position of the x coordinate of the observed pixel in a tile domain of the captured image; 
 calculating a second relative position of one of a plurality of dots in a tile domain of a reference image using the extracted class identifier; and 
 calculating a disparity of the observed pixel using the first relative position and the second relative position. 
 
     
     
       2. The method of  claim 1 ,
 wherein the calculating of the first relative position includes: 
 translating the x coordinate of the observed pixel into the tile domain of the captured image to generate the first relative position in the tile domain of the captured image. 
 
     
     
       3. The method of  claim 1 , further comprising:
 generating a projected image to be projected onto an object from the reference image, wherein the reference image includes a plurality of tiles, 
 wherein each of the tiles includes the plurality of dots each of which is assigned to one of a plurality of class identifiers and wherein the extracted class identifier is one of the plurality of class identifiers; and 
 generating the captured image from the projected image onto the object. 
 
     
     
       4. The method of  claim 3 ,
 wherein the projected image is projected as infrared light. 
 
     
     
       5. The method of  claim 3 ,
 wherein the plurality of class identifiers each is assigned to one of the plurality of dots left to right and row-by-row within the tile domain of the reference image, 
 wherein the plurality of class identifiers have continuously-increasing values from a first corner of the tile domain of the reference image to a second corner of the tile domain of the reference image facing diagonally the first corner. 
 
     
     
       6. The method of  claim 3 ,
 wherein the calculating of the second relative position is performed by using a function of which an input includes the extracted class identifier and an output is the second relative position, and wherein the function represents a distribution of the plurality of class identifiers on the tile domain of the reference image. 
 
     
     
       7. The method of  claim 6 ,
 wherein the input of the function further includes the y coordinate of the observed pixel. 
 
     
     
       8. The method of  claim 7 ,
 wherein the plurality of dots of the reference image is rotated at a predefined rotation angle in the projected image, and 
 wherein the input of the function further includes the predefined rotation angle. 
 
     
     
       9. The method of  claim 8 ,
 wherein the input of the function further includes a dot sampling rate, and 
 wherein the dot sampling rate corresponds to a number of pixels horizontally arranged in each of the plurality of dots. 
 
     
     
       10. The method of  claim 3 ,
 wherein each of the plurality of dots is divided into a plurality of sub-dots and Wherein the plurality of class identifiers are assigned to the plurality of sub-dots. 
 
     
     
       11. A method of reconstructing a three dimensional image using a structured light pattern system, the method comprising:
 generating a projected image to be projected onto an object from a reference image, wherein the reference image includes a plurality of tiles, wherein each of the tiles includes a plurality of dots each of which is assigned to one of a plurality of class identifiers; 
 generating a captured image from the projected image onto the object; 
 extracting a class identifier of an observed pixel on the captured image; and 
 calculating a first relative position of the observed pixel in a tile domain of the captured image; 
 calculating a second relative position of one of the plurality of dots in a tile domain of the reference image using the extracted class identifier and a y coordinate of the observed pixel; and 
 calculating a disparity of the observed pixel using the first relative position and the second relative position. 
 
     
     
       12. The method of  claim 11 ,
 wherein the plurality of class identifiers each is assigned to one of the plurality of dots loft to right and row-by-row within the tile domain of the reference image, 
 wherein the plurality of class identifiers have continuously-increasing values from a first corner of the tile domain of the reference image to a second corner of the tile domain of the reference image facing diagonally the first corner. 
 
     
     
       13. The method of  claim 11 ,
 wherein the calculating of the first relative position includes: 
 translating the observed pixel into the tile domain of the captured image to generate the first relative position in the tile domain of the captured image. 
 
     
     
       14. The method of  claim 11 ,
 wherein the calculating of the second relative position is performed by using a function of which an input includes the extracted class identifier and the y-coordinate of the observed pixel and an output is the second relative position, and wherein the function represents a distribution of the plurality of class identifiers on the tile domain of the reference image. 
 
     
     
       15. The method of  claim 14 ,
 wherein the plurality of dots of the reference image is rotated at a predefined rotation angle in the projected image, and 
 wherein the input of the function further includes the predefined rotation angle. 
 
     
     
       16. A method of reconstructing a three dimensional image using a structured light pattern system, the method comprising:
 generating a reference image including a plurality of tiles, wherein each of the tiles includes a plurality of dots each of which is assigned to one of a plurality of class identifiers; 
 projecting a projected image generated from the reference image onto an object having a three-dimensional surface; 
 capturing the projected image onto the object to generate a captured image; 
 extracting a class identifier from an observed pixel on the captured image; and 
 translating the observed pixel into a tile domain of the captured image to generate a first relative position in the tile domain of the captured image; 
 calculating a second relative position of one of the plurality of dots in a tile domain of the reference image using the extracted class identifier; 
 calculating a disparity of the observed pixel using the first relative position and the second relative position; and 
 generating a three-dimensional profile of the object using the disparity. 
 
     
     
       17. The method of  claim 16 ,
 wherein the calculating of the second relative position of the reference image is performed using a function representing a distribution of the plurality of class identifiers on the tile domain of the reference image, 
 wherein the extracted class identifier is a first input of the function, and 
 wherein a row number of the observed pixel in the captured image is a second input of the function. 
 
     
     
       18. The method of  claim 17 ,
 wherein the plurality of dots of the reference image is rotated at a predefined rotation angle in the projected image, and 
 wherein the predefined rotation angle is a third input of the function. 
 
     
     
       19. The method of  claim 17 ,
 wherein a dot sampling rate is a fourth input of the function, and 
 wherein the dot sampling rate corresponds to a number of pixels horizontally arranged in each of the plurality of dots. 
 
     
     
       20. The method of  claim 16 ,
 wherein the generating of the reference image includes: 
 assigning each of the plurality of class identifiers to one of the plurality of dots, 
 wherein integer values of the plurality of class identifiers increase successively.